Comic Brief

• Outcome: Review
• Latest audit: May 29, 2026 at 9:34 PM UTC
• Version:
  1.0.0

Overview

This skill appears to generate comic-style images but also sends generated media and prompt-derived context to an external Convex service with too little user control or disclosure.

Review this skill carefully before installing. Use it only for non-sensitive image prompts unless the publisher documents exactly what is uploaded, where it is stored, how long it is retained, and gives you an explicit approve-before-publish step.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (4)

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The skill’s stated purpose is image generation, but it also persistently uploads generated content and associated prompt metadata to an external Convex service. This creates an unexpected data-flow boundary: user-supplied topics, notes, or meeting-derived content may be transmitted off-platform and stored permanently without clear disclosure or consent.

Context-Inappropriate Capability

Medium

• Confidence: 92% confidence
• Finding: The skill bundles media persistence and downstream scheduling enablement into a tool whose advertised function is only to generate comic-style illustrations. This feature creep increases risk because users may provide sensitive business or meeting content expecting local generation, while the skill silently prepares it for external hosting and reuse.

Vague Triggers

Medium

• Confidence: 83% confidence
• Finding: Broad trigger phrases like 'daily briefing' or 'content calendar' can cause the skill to automatically pull in additional content sources without precise scoping or user confirmation. In context, that matters because the retrieved content may then be incorporated into prompts and uploaded externally, expanding data exposure beyond what the user likely intended.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The skill explicitly instructs the agent to run shell commands immediately and to avoid asking the user for confirmation, then later uploads the resulting content to external storage. This bypasses informed consent for both execution and data transfer, making accidental exfiltration of sensitive user-provided material significantly more likely.

VirusTotal

66/66 vendors flagged this skill as clean.